DESCRIPTION (Adapted from the Applicant?s Abstract): Objectives: This is one of four identical revised applications of a collaborative R01 proposal being conducted at the University of Washington, Seattle, WA, University of North Carolina (UNC), Chapel Hill, NC, Harvard Medical School, Boston, MA and Case Western University, Cleveland, Ohio. The study will examine the long-term effectiveness of three different antipsychotic medications in the treatment of early onset schizophrenia and schizoaffective disorder: risperidone (RIS), olanzapine (OLA), the molindone (MOL) over a one year period. Specific Aims: 1) To determine the efficacy of MOL, RIS, and OLA in reducing psychotic symptoms in youth with schizophrenia and related disorders; 2) To determine the ability to sustain treatment over one year with each of these agents; 3) To examine the effects of treatment on adaptive and neurocognitive functioning; and 4) To examine the safety and tolerability of these agents in the pediatric population, particularly potential effects on neuropyramidal symptoms and weight gain. Research Design: 168 youths (ages 8-19 years) with schizophrenia, schizophreniform disorder or schizoaffective disorder and active psychotic symptoms will be recruited across four sites. Subjects will be randomly assigned to double-blind treatment with one of the three study medications. Standard dosage schedules will be followed, with modifications allowed dependent on the clinical status of subjects. Antipsychotic agents will be cross-tapered over the first week of the study to prevent exacerbation of psychotic symptoms. Symptom ratings and neurocognitive testing will be performed at baseline and repeated at specific intervals. The acute phase of treatment will last 8 weeks. Subjects with clinically significant improvement and without intolerable side effects, will continue maintenance therapy for an additional 44 weeks. Tolerance of the study medications will be systematically monitored with assessments for extrapyramidal side effects and weight gain. Revision of treatment algorithms, reliability testing and data analysis will be coordinated between the four sites. Randomization, preparation of study medications, and overall management of the database will be centralized at UNC. Revisions: To address reviewers concerns, the revised proposal has added a fourth site (Harvard), revised the data management and analysis plan, and adjusted some of the primary outcome measures. Significance: There are very few controlled studies to inform clinical practice for the treatment of youth with psychotic disorders. This study will provide information about the comparative effectiveness of the most commonly used and representative antipsychotic drugs in youth with schizophrenia spectrum disorders.